Adolescent Offending and the Segregation of Poverty in Urban Neighbourhoods and Schools: An Assessment of Contextual Effects from the Standpoint of Situational Action Theory

Table 4

Blockwise cross-nested multilevel models.

Dependent variable/serious offending scale	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7
Fixed effects	Unstandardised slopes of standardised parameters
Structural background variables at level 1
Sex (Boys coded 1)			0.173	0.140	0.117	0.095	0.106
Immigrant background (both parents native coded 0)			0.085	0.055	0.036	0.056	0.060
One-parent family (one-parent family coded 1)			0.010	−0.000	0.000	−0.010	−0.01
Household disadvantage (disadvantage coded 1)			0.015	0.009	0.011	0.006	0.019
Level 2 characteristics
Neighbourhood-level disadvantage		0.035	0.018	0.015	0.013	0.014	0.020
School-level disadvantage		0.137	0.107	0.097	0.092	0.072	0.072
Social mechanisms at level 1
School commitment				−0.111	−0.053	−0.013	−0.014
Parental monitoring				−0.230	−0.179	−0.116	−0.105
Propensity					0.181	0.107	0.099
Exposure to criminogenic settings						0.268	0.186
Propensity* exposure to criminogenic settings							0.212
Random effects
Unique individual-level variance	1.028	1.030	0.996	0.913	0.890	0.843	0.799
Unique neighbourhood-level variance	0.002	0.000	0.000	0.000	0.000	0.000	0.001
Unique school-level variance	0.029	0.005	0.002	0.004	0.003	0.001	0.000
Sign: deviance test of model fit improvement	—	*	*	*	*	*	*
Decomposition pseudo-𝐑^𝟐^27
Neighbourhood level (%)	—	95	96	91	93.5	70.56	45
School level (%)	—	79.56	92.18	85.72	87.04	95.39	97.80
Individual level (%)	—	—	3.02	11.12	13.35	17.94	27.34

Note. Bold parameters are statistically significant (𝑃<.05 or better). All of the variables have been standardised in order to gain insight into the standardised effects of the regression coefficients.
—: Not applicable (not calculated in the model).
*: Significant improvement in model fit at 𝑃<0.05.